You're interviewing a candidate and they mention juggling work with their kid's baseball practice. Suddenly: warm, fuzzy recognition — you do that too. Craig della Penna opens Cipher's treatment of similarity bias with exactly that moment, and with the trap inside it: "Just because you're awesome, and they like to do the things you want to do, doesn't mean they're awesome too when it comes to what is needed to do the job."
Similarity-attraction bias is the pull to rate people more highly the more they resemble us — school, background, age, industry, favourite restaurant, owning a dog. It's not malice; it's wiring. And it costs teams in three specific ways.
1. You hire echoes, and the goals needed range
A team selected for resemblance to its manager duplicates its manager's strengths — and its manager's gaps, now with nobody positioned to cover them. The business case for the alternative is quantified: companies in the top quartile for gender diversity on executive teams were 21% more likely to out-profit their peers (McKinsey, as cited in Cipher, ch. 11).
2. You misread "culture fit," and it filters the wrong people
Fit gets invoked as if it meant shared values. Under similarity bias it silently becomes shared background — and starts filtering out the candidate from a different industry whose accomplishments you discount, or the one who worked for a competitor. The Yale study the book cites shows how deep this runs: scientists trained to be objective still rated men more competent than identically-qualified women, and priced them $4,000 higher. (As cited in Cipher, ch. 11.) Nobody opts into bias; that's what makes it bias.
3. Your first ten minutes decide the next fifty
The moment of recognition — same school, same sport — arrives early, and the rest of the interview becomes a search for confirmation. The friendly, unstructured conversation that follows produces what Cipher calls "the statistically proven worst results possible."
The fix is structural, not moral
You can't train the pull away — the book is blunt that bias-free humans aren't on offer. What works is what removes similarity from the decision path:
- Define the standard first. A Cipher — what the role must deliver — before anyone interviews. When "fit" means fit to the deliverables, the baseball practice stops being data.
- Ask every candidate the same mapped questions. "If you interview three people, and ask them all different questions, how do you know which one is best?" (Cipher, ch. 11.)
- Rate evidence against the standard, independently, before discussing. The debrief compares scored evidence, not warmth.
During the interview itself, HireBest's AI Coach watches for the drift this bias produces — the leading question, the tangent that's really rapport-building, the deliverable never probed because the conversation was too pleasant to interrupt.
"The best hiring decisions come from having a Cipher designed in advance, and following a structured interview process focused on data, not feelings." — Cipher, ch. 11
See the AI Coach catch it live →→ /product
---
Frequently asked questions
What is similarity-attraction bias?
The tendency to evaluate people more favourably the more they resemble you — background, education, interests, life stage. In hiring it inflates ratings for candidates who feel familiar and discounts those who don't, independent of ability to do the job.
Is hiring for culture fit the same as similarity bias?
It becomes it when "fit" is judged by feel. Fit to explicit, written values or role requirements is legitimate; fit as unexamined resemblance is the bias with a friendlier name.
How do you reduce similarity bias in interviews?
Take resemblance out of the decision path: define the role's deliverables first, ask every candidate the same mapped questions, and rate written evidence independently before any group discussion.
Sources
- Hunt, D.V. et al. “Delivering through Diversity.” McKinsey & Company, 2020 — as cited in Cipher, ch. 11.
- Agarwal, P. “Here Is How Bias Can Affect Recruitment in Your Organisation.” Forbes, 2018 — as cited in Cipher, ch. 11.
- della Penna, C. Cipher: A Proven Framework to Hire the Best People. Third Act Publishers, 2026.